Psychometric properties of the Prenatal Health Behavior Scale in mid- and late pregnancy.

INTRODUCTION: Psychometrically sound self-report measures are important tools to advance research on health behaviors in pregnancy. The Prenatal Health Behavior Scale (PHBS) has been used in prior studies to quantify health behaviors and examine their associations with relevant variables, but its psychometric properties have not been carefully investigated.
METHODS: In a sample of low-risk women, we examined the factor structure, reliability, and validity of the PHBS at mid- (N = 138) and late (N = 102) pregnancy.
RESULTS: In confirmatory factor analysis, items on the PHBS loaded onto a single reliable factor, but separating health-promoting and health-impairing behavior into two factors had better model fit. The scales defined by these two factors had good internal consistency in mid- and late pregnancy, were only moderately correlated, and were predicted by somewhat different sets of sociodemographic and psychological variables. Pregnancy-specific stress was a robust predictor of health-promoting and health-impairing behavior across time. DISCUSSION: Results bolster confidence in the reliability and validity of the PHBS and its appropriateness as a self-report tool for investigation of health behaviors among pregnant women. This study is one of the first investigations to examine patterns and predictors of health behavior practices at two time periods of pregnancy.